Climate is the expression of the normal ranges and dominant, repeated patterns of temperature, rainfall and associated radiation, humidity and wind regimes that prevail at any given location. Most approaches to describing climate rely on the analysis of long-run means of monthly, seasonal or annual climatic variables. The Koeppen schema and FAO's Agro-Ecological Zones (AEZ) approach are the most widely-used climate classification approaches. HarvestChoice makes extensive use of the AEZ method in the qualitative assessment of, for example, the suitability of geographic areas for the production of specific crops or for the application of specific production interventions.

Weather, by contrast, reflects the short-term variation of climate variables in both time and space, typically monitored on an hourly or daily basis. Weather-related phenomena include extreme events such as rainstorms, heat waves, frost, and droughts. In HarvestChoice, we use weather data (typically time series of monthly or daily variables) extensively in simulating the growth of crops (using cropping system growth simulation models such as DSSAT and APSIM that operate on a daily time step) as well as in assessing the potential prevalence and persistence of biotic stresses (using the CLIMEX model).

HarvestChoice Primary Climate/Weather Data Sources

University of East Anglia – Climate Research Unit’s CRU TS  

  • Gridded historical global climate database, interpolated from climate observations from the global network of meteorological stations
  • Various releases; currently CRU-TS v3.1 covering 1901-2009 monthly grids at 0.5 degree
  • Variables (monthly mean): cloud coverage, diurnal temperature range, frost day frequency, precipitation, daily mean temperature, monthly average daily minimum and maximum temperature, vapor pressure, wet day frequency, and potential evapotranspiration. Count of stations used to estimate the variable in each grid cell is also provided.

University of Delaware –  Willmott, Matsuura and Collaborators’ Global Climate Resource Data  

  • Gridded historical global climate database, interpolated from meteorological stations and oceanic grid nodes. Number of stations used in the interpolation process was 24,941 for air temperature and 26,858 for precipitation, respectively
  • Current release (August 2009) covers 1900-2008 monthly grids at 0.5 degree
  • Variables (monthly mean): Average air temperature, precipitation, terrestrial water budgets, and moisture indices

NASA POWER Agroclimatology Database

  • Satellite and model-derived daily solar radiation and meteorological data
  • Global coverage on 1 degree grids. Different temporal coverage depending on variable; complete coverage since 1997
  • Variables (daily): Solar radiation, dew point, air temperature (max and min), and precipitation

Daily weather data generated from MarkSim

  • Stochastically generated daily weather data based on the analysis of historical observations from meteorological stations and Markov chain method
  • Current version covers most tropics area (Latin America, Africa, and Asia) with the spatial resolution of 2.5 arc-minute for Asia and 10 arc-minute for Latin America and Africa
  • Variables (daily): Solar radiation, temperature (max and min), and precipitation.

Downscaled Climate Projection Data by Climate Change, Agriculture, and Food Security (CCAFS)

  • Monthly climate profile of future time slices (2030s, 2050s, and 2080s) generated by spatial downscaling of General Circulation Models. Climate models and scenario runs are based on 2007's Fourth Assessment Report of the Intergovernmental Panel on Climate Change (IPCC)
  • Global coverage on 5 arc-minute grids
  • Variables (monthly mean): Solar radiation, temperatures (min and max), precipitation, and the number of rainy days per month.

Topic Contacts